FAQs (Data Related) Case Definition
Note: These questions refer to both NHIS and BRFSS surveys, unless otherwise indicated.
- What is the current case definition for self-reported arthritis?
- Why does CDC include the four types of arthritis to the end of the survey question addressing doctor-diagnosed arthritis: “Have you ever been told by a doctor or other health professional you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?”
- Are self-reports for arthritis valid?
- Which estimate best indicates the prevalence of arthritis for the nation?
- Is self-reported information on the type of arthritis valid?
- What changes are expected for future BRFSS surveys?
- What does the CDC Arthritis Program recommend state arthritis programs do regarding the BRFSS survey in even years?
- Is there more than one way to define a “case” of arthritis?
- Why did CDC decide to drop questions on chronic joint symptoms and cease surveillance of “possible arthritis”?
Since 2002, CDC, partners, and outside experts have agreed on a case definition of simply “doctor-diagnosed arthritis.”
Both the Behavioral Risk Factor Surveillance System and the National Health Interview Survey use the following question to identify adults with arthritis “Have you ever been told by a doctor or other health professional that you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?
For more information on the evolution of the case definition read more.
Why does CDC include the four types of arthritis to the end of the survey question addressing doctor-diagnosed arthritis: “Have you ever been told by a doctor or other health professional you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?”
For public health surveillance, we are trying to capture more than 100 diseases that are considered “arthritis or other rheumatic conditions.” Some of these conditions may not be readily recognized as arthritis by survey respondents. Experts recommended adding these conditions.
Self-reports have been shown to be valid for surveillance purposes.1
National estimates should come from a national survey; CDC uses the National Health Interview Survey (NHIS) for such estimates. Healthy People 2020 also recommends using NHIS data for national estimates.
Using NHIS data (2010-2012), we estimate that 52.5 million (22.7% ) adults had doctor-diagnosed arthritis. 2
The CDC Arthritis Program strongly discourages the collection and reporting of self-reported data on arthritis type (e.g., osteoarthritis, rheumatoid arthritis, lupus) because self-reported condition data have been shown to be inaccurate in studies done by program staff and other researchers. For example, a study conducted by the Agency for Healthcare Quality and Research examined examining how frequently the arthritis type reported by participants in the Medical Expenditure Panel Survey (MEPS) agreed with each of the participants’ doctor’s diagnosis. This study found that among persons whose doctor had diagnosed them with osteoarthritis, only 12% of survey participants reported that they had osteoarthritis. Since self-reported information on type of arthritis is not valid, the CDC arthritis program discourages states and researchers from measuring and using data on self-reported arthritis condition type. (Click here for other such references.)
- The Medical Expenditure Panel Survey is a nationally representative survey of the U.S. civilian non-institutionalized population.
Starting in 2003, the BRFSS Arthritis Core (Arthritis Burden Questions) are offered on the core survey in odd numbered years. In 2011, the case finding question for doctor-diagnosed arthritis was moved out of the Arthritis Burden module. State programs are encouraged to add the optional module known as the BRFSS Arthritis Optional Module (Arthritis Management Questions) in the same odd-numbered year. Although states may be interested in using the questions every year, CDC will NOT financially or analytically support the use of any arthritis questions in even numbered years. States may add them as state-added questions.
CDC is not supporting the use of arthritis questions in even years. This means we are not funding the BRFSS to put the questions on the survey nor are we providing our standard analytic tables, as we do for all states using odd number year data. Any use of these questions in even years would have to be as state-added questions. We do not feel it is necessary to have the data each year, as the prevalence is unlikely to change rapidly and with lots of programs competing for space on the questionnaire, one can quickly wear their welcome out with the BRFSS coordinator. In infrequent instances, when a state feels it is imperative for its program to make substate prevalence estimates (e.g., for large cities), it may choose to use the arthritis questions in even years. CDC is unable to provide any support for doing this and we don’t recommend it.
We expect the BRFSS arthritis case definition question to be asked annually through 2015, at which time the Survey will reconsider all questions in the Core Chronic Health Conditions Module. We expect the remaining four questions of the Arthritis Burden Module to be asked in in all states in odd number years. A list of BRFSS Arthritis Core (Arthritis Burden Questions) by year is available.
- Is there more than one way to define a “case” of arthritis?
Yes, the purpose (e.g., estimating prevalence or assessing burden) and source of the data determine the arthritis case definition to be used. In general, data used in arthritis surveillance falls into two broad categories, self-reported data and health care system data.
For the purpose of estimating the population prevalence of arthritis, we recommend using self-reported data as the data source and using doctor-diagnosed arthritis as the case definition. Both the state-based Behavioral Risk Factor Surveillance System (BRFSS) and National Health Interview Survey (NHIS) are self-reported data surveys, and use doctor-diagnosed arthritis as the case definition. A “case” of doctor-diagnosed arthritis in these surveys is defined as a “yes” answer to the following question: “Have you EVER been told by a doctor or other health professional that you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?” Further information on the CDC arthritis case definition is available.
For assessing the burden of arthritis using health care system data (e.g. hospital discharge data, ambulatory care, death.) a “case” of arthritis is defined by the National Arthritis Data Workgroup definition using arthritis related ICD–9–CM codes [52K] and is labeled “arthritis and other rheumatic conditions”. The CDC discourages states from pursuing these assessments unless there is a direct link to program activity that is consistent with the state arthritis plan.
Be aware that other studies (e.g., clinical trials) including some statistics and studies cited elsewhere on this Website, have different data sources and use different arthritis case definitions.
The state-based Behavioral Risk Factor Surveillance System (BRFSS) included 2 questions on joint symptoms between the years of 2002 and 2007. “Possible arthritis” is defined as a person without doctor-diagnosed arthritis who answers “yes” to both of the following questions: “During the past 30 days, have you had any symptoms of pain, aching or stiffness in or around a joint?” and “Did your joint symptoms first begin more than 3 months ago?” CDC conducted an expanded data collection in 4 states in 2005 to learn more about people with “possible arthritis.” Results indicate that over 86% of persons with possible arthritis who had seen a doctor did not appear to have arthritis and that the majority of those who had not seen a doctor did not deem their symptoms serious enough for medical intervention. The CDC arthritis program reasons that persons with possible arthritis would not be amenable to public health programmatic interventions for arthritis and that maintaining surveillance of this group is currently not necessary.
Further analysis of BRFSS data on “possible arthritis” indicated the group has a different profile than those with doctor diagnosed arthritis. Unlike doctor-diagnosed arthritis, “possible arthritis” decreases with age and is more common in men than women. A smaller proportion of people with possible arthritis report activity limitation due to joint symptoms compared to those with doctor-diagnosed arthritis. In 2009 the 2 joint symptoms questions were dropped from the BRFSS. These questions continue to be collected through the National Health Interview Survey (NHIS).
- Sacks JJ, Harrold LR, Helmick CG, Gurwitz JH, Emani S, Yood RA. Validation of a surveillance case definition for arthritis. J Rheumatol 2005;32:340–347
- Barbour KE, Helmick CG, Theis KA, Murphy LB, Hootman JM, Brady, Cheng YJ. Prevalence of doctor-diagnosed arthritis and arthritis-attributable activity limitation—United States, 2010-2012. MMWR 2013;62(14):869-873.
- Star VL, Scott JC, Sherwin R, Lane N, Nevitt MC, Hochberg MC. Validity of self-reported rheumatoid arthritis in elderly women. J Rheum 1996;23:1862-5.
- Machlin S, Cohen J, Elixhauser A, Beauregard K, Steiner C. Sensitivity of household reported medical conditions in the medical expenditure panel survey. Med Care. 2009 Jun;47(6):618–25.
- Brinton LA, Buckley LM, Dvorkina O, Lubin JH, Colton T, Murray MC, Hoover RN. Risk of connective tissue disorder among breast implant patients. Am J Epidemiol 2004;160:619-27.
- March LM, Schwarz JM, Carfrae BH, Bagge E. Clinical validation of self-reported osteoarthritis. Osteoarthritis and Cartilage 1998;6:87-93.
- Kvien TK, Glennas A, Knudsrod OG, Smedstad LM, The validity of self-reported diagnosis of rheumatoid arthritis: results from a population survey followed by clinical examinations. J Rheum 1996;23:1866-71.
- Bolen J, Helmick CG, Sacks JJ, Gizlice Z, Potter C. Should People Who Have Joint Symptoms, But No Diagnosis of Arthritis from a Doctor, Be Included in Surveillance Efforts? Arthritis Care & Research 2011;63(1):150-154.